Dear cluster users,

hpc.nrw and MathWorks cordially invite you to three additional MATLAB workshops. Registration is now open for the following workshops:

• Interoperability of MATLAB and Python on June 30, 2022
• Machine Learning with MATLAB on July 7, 2022
• Deep Learning with MATLAB on July 14, 2022

To register for one of the courses, please click on the corresponding link.

The courses will be held online via Zoom and go from 2 PM to 5 PM on the respective day. Due to the high interest we will randomly select the participants from all registrations. You will know about 2 weeks before the course whether you were accepted.

Since MATLAB Online will be used during the exercises, you do not need a MATLAB installation to participate.

If you have questions, you can reach the event organizer under jan.steiner@uni-siegen.de.

Best regards

Your HPC Team

Update May 12: the problem has been fixed. New jobs can be queued. Old jobs have restarted.

Original text:

Dear cluster users,

due to a technical problem it is currently not possible to queue new jobs on OMNI. The status of the running jobs is still unclear.

We will notify you when the problem has been fixed.

Best regards

Your HPC Team

Dear cluster users,

on May 17 and 18 we will offer our next course “MATLAB – Parallelization and Performance”. The course will take place over two consecutive evenings. It will be held in German.

If you would like to make your MATLAB code faster, if you are running into problems with insufficient memory or if you would simply like to learn all the ways you can use MATLAB on the cluster, then this course is for you.

The course will be held online via Zoom, regardless of how the coronavirus situation develops. All registered participants will receive the Zoom meeting information shortly before the course starts.

If you are interested in participating despite the course being in German, please refer to this page.

Best regards

Your HPC Team

Dear cluster users, 

there will not be a consultation hour on April 12, nor on April 19, 2022. If you have any questions, it is best to send an email to hpc-support@uni-siegen.de.

Best regards

Your HPC Team

Dear cluster users,

in the upcoming summer semester 21/22 we offer once again our C++ introduction course. It will be held by our team member Verena Betge and will cover the basics of programming in general and in C++ specifically. The course is intended for absolute programming newbies. The course will be held in German.

Unlike our other courses, this course will not be held in a single block, but rather distributed over the semester’s lecture period. The first date is April 11, 2022, there will be 10 blocks overall. The blocks will be held every Monday 10:15-11:45.

The course is available to all students and employees. There is no prerequisite knowledge.

If you are interested, please send an email to verena.betge@uni-siegen.de.

For more information about topics and dates (in German), please go to the course page: https://cluster.uni-siegen.de/kurs/kurs-c-einfuehrung-deutsch-woechentlich-ab-11-april-2022/

Best regards

Your HPC Team

Dear cluster users,

on the morning of March 3, 2022, the file systems /work and /fast will be unavailable due to maintenance work. The downtime is necessary because network components need to be replaced.

Jobs that are in the waiting queue at that point will only start after the maintenance. Jobs that cannot complete before the maintenance (e.g. a 20-day job queued 19 days before the maintenance) will be held until after the maintenance.

Logging in should still be possible.

The maintenance work should be completed around noon.

Best regards
Your HPC Team

Dear cluster users,

due to a power outage in Siegen-Weidenau yesterday (Wednesday, Feb. 2, 2022) the compute nodes of OMNI shut themselves down. The HPC nodes, GPU nodes and SMP nodes were affected by this. They were restarted manually shortly after that.

Jobs that were running at the time were killed. These jobs should have restarted by themselves. Jobs that were in the waiting queue should be unaffected.

We highly recommend that you check all your jobs and requeue them if necessary.

Best regards

Your HPC Team

Dear cluster users,

due to the unusually high demand this semester, we will offer another cluster introduction course. The course will take place over two consecutive evenings, namely February 15 and 16, 2022. It will be held in English.

If you are using the OMNI cluster or other compute resources of Uni Siegen, this course is your most important entry point. We recommend that every cluster user attend this course at least once.

For more details and registration information, please refer to this page.

If you would like to participate, simply send an email to jan.steiner@uni-siegen.de.

Dear cluster users,

the next Linux introduction course will be held in German. If you wish to participate anyway, switch to the German version of this page for more details.

Best regards

Your HPC Team

Dear cluster users,

You are cordially invited to the 2-day course “GPU Computing at HPC.NRW”, organized by the competence network HPC.NRW and Paderborn Center for Parallel Computing (PC²).

Date and time: Jan. 18 – 19, 2022 (9:00 – 17:00 on each day)

The registrations are open to everyone at
https://events.uni-paderborn.de/e/gpu2022

GPU Computing plays a significant role in High-Performance Computation. This 2-day course on GPU Computing covers the basic concepts of GPU architectures, the CUDA C/C++ programming as well as GPU code profiling and optimization. The topics included are:

Day 1: Introduction to GPU computing and CUDA
Day 2: Advanced CUDA, optimization and profiling

Each day of the course consists of approximately 4 hours of lectures and 3 hours of hands-on exercises. The aim is that by the end of this 2-day course you are able to write CUDA programs on GPU and tune the GPU code performance by using the profiling and optimization tools from NVIDIA.

Best regards,
Your HPC Team